Add to Favourites
To login click here
This study presents IRON, an integrative radiogenomic framework to predict the volumetric response of heterogeneous, multi-site ovarian cancer to NACT. The framework uses two independent, highly annotated data sets including clinical, chemotherapy treatment, CA-125, ctDNA, and radiomics features extracted from all primary and metastatic disease at diagnosis. The features are used as an input to an ensemble machine learning model to predict volume shrinkage during NACT. The approach was validated on an external training data set and demonstrated that radiomics features are essential to obtain significant predictive power.